Research map / Video / Video segmentation and tracking
Hierarchical Visual Prompt Learning for Continual Video Instance Segmentation
Jiahua Dong, Hui Yin, Wenqi Liang, Hanbin Zhao, Henghui Ding, Nicu Sebe, Salman Khan, Fahad Shahbaz Khan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add task-specific frame prompt (TFP) to base (ablation) | HVPL · YouTube-VIS 2021 30-10 | AP: 27.53 -> 30.02 | Improved |
| add task-specific video prompt and video context decoder (TVP) (ablation) | HVPL · YouTube-VIS 2021 30-10 | AP: 30.02 -> 39.07 | Improved |
| add graph-guided state space (GSS) layers (ablation) | HVPL · YouTube-VIS 2021 30-10 | AP: 39.07 -> 40.37 | Improved |
| add orthogonal gradient correction (OGC) (ablation) | HVPL · YouTube-VIS 2021 30-10 | AP: 40.37 -> 43.24 | Improved |
| add all HVPL modules to base (forgetting rate) (ablation) | HVPL · YouTube-VIS 2021 30-10 | FAP: 96.48 -> 15.69 | Improved |
| add TFP+TVP to base (forgetting rate) (ablation) | HVPL · YouTube-VIS 2021 30-10 | FAP: 74.88 -> 31.05 | Improved |
| add OGC (forgetting rate) (ablation) | HVPL · YouTube-VIS 2021 30-10 | FAP: 30.40 -> 15.69 | Improved |
| HVPL instead of ECLIPSE | HVPL vs ECLIPSE · YouTube-VIS 2021 30-10 | AP: 29.68 -> 43.24 | Improved |
| HVPL instead of ECLIPSE | HVPL vs ECLIPSE · OVIS 15-5 | AP: 6.62 -> 11.09 | Improved |
| HVPL instead of ECLIPSE | HVPL vs ECLIPSE · YouTube-VIS 2019 20-2 | AP: 30.24 -> 31.68 | Improved |
| HVPL instead of ECLIPSE at last task t=6 | HVPL vs ECLIPSE · YouTube-VIS 2021 20-4 | AP: 22.76 -> 22.81 | No clear effect |
| vary elastic threshold xi 0.2 to 1.0 (ablation) | HVPL · YouTube-VIS 2021 30-10 | AP: 0.2 to 1.0 | No clear effect |
Closest papers by meaning
- Improving Continual Video Instance Segmentation via Spatial-Temporal Balanced Mixture-of-Experts Adapters (NeurIPS 2026 · similarity 0.89)
- End-to-End Video Instance Segmentation With Transformers (CVPR 2021 · similarity 0.88)
- A Generalized Framework for Video Instance Segmentation (CVPR 2023 · similarity 0.88)
- Causal Prompts for Open-Vocabulary Video Instance Segmentation (TPAMI 2026 · similarity 0.88)
- Hybrid Instance-Aware Temporal Fusion for Online Video Instance Segmentation (AAAI 2022 · similarity 0.88)
- STOP: Integrated Spatial-Temporal Dynamic Prompting for Video Understanding (CVPR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
